 original innovation
 Open Access
 Published:
Investigation of the performance of grouted couplers in vehicle impacted reinforced concrete ABC bridge piers
Advances in Bridge Engineering volumeÂ 3, ArticleÂ number:Â 18 (2022)
Abstract
Increased dynamic impact on bridge piers caused by seismic events, blasts, and vehicular impact have become increasingly common. Recent research efforts indicate that code provisions for designing reinforced concrete members to withstand such dynamic loads are inadequate and need additional insights for this purpose. Numerous works have been undertaken to investigate reinforced concrete (RC) traditional bridge pier performance on high strain rate loading. However, little attention has been given to evaluate the performance of connections used in present day bridges including accelerated bridge constructions (ABC) to withstand vehicle impacts, and hence, is relatively unknown. In this study, the use of grouted couplers to contain the unbalanced moments resulting from vehicular impact forces exceeding the moment capacity of the reinforced concrete piers and avoiding extensive damage to the piers is investigated. A representative column, typical of those specified by state departments of transportation, is studied to determine the performance. The performance of the coupler is investigated for both dynamic and static combined stresses. Quasistatic to dynamic strain rates of steel reinforcement connected to the couplers is also evaluated. Quantifying the stresses and strains developed at coupler region from dynamic impact can help coupler manufacturers to optimize the strength properties, thus improving serviceability. This study investigated utilizing splice sleeves in mitigating the formation of plastic hinges, as well as addressing the essential properties of coupler sections required to adequately carry out this function, and will provide a useful design tool for the manufacturers, forensic structural engineers, and practitioners.
1 Introduction
Bridge piers experience highly dynamic loading due to a variety of loading conditions including seismic events, blast events, and vehicular collisions. This may cause health deterioration of the column ranging from cosmetic damage to collapse. The other mechanisms comprising of dynamic impact loads (vehicular collisions and blast) have received more attention for traditional RC bridge piers. However, the performance of connections used in present day bridges including accelerated bridge constructions (ABC) to withstand high strain rate loading from vehicle impacts is relatively unknown. Despite different studies showing that vehicular collisions with bridge elements are the most common dynamic impact scenario, especially with the increasing volumes of vehicular traffic (Sharma et al. 2015). As such, it is imperative to investigate the response of reinforced concrete bridge (RC) piers to vehicle collisions and also possible methods of enhancing the capacity of the piers to withstand such events.
Investigations into the failure mechanisms of bridge piers subjected to vehicular impact have shown that the failure of these members will primarily occur from excessive shear as well as overturning moments occurring at the column base (ElTawil et al. 2005; Thilakarathna et al. 2010). Recent studies of the structural reliability of RC columns subjected to sequential loading of blast and vehicular impact have shown that the structural reliability of the columns are particularly sensitive to the shear reinforcement, column diameter, and reinforcement ratio (Thomas et al. 2018; Roy et al. 2022). This implies that an increase in the stiffness of the column could possibly help it withstand the external forces caused by vehicular impact. Traditional pier sections have been studied as a means of increasing the stiffness of reinforced concrete (RC) columns to improve their seismic performance, energy dissipation (Roy and Sorensen 2021a), failure mechanism in terms of crack propagation (Roy and Sorensen 2021b), and to overcome the development of plastic hinges (Ebrahimpour et al. 2016; GirÃ£o Coelho et al. 2012; Pantelides et al. 2014; Tazarv and Saiidi 2016). The failure mechanism of the grouted coupler at the post behavior from vehicle impact at high strain rate loading has been presented in this study. However, the effect of these couplers performance on the response of these piers to vehicle impact is still relatively unknown, and hence needs an additional scrutiny.
This study presents the grouted coupler section undergoing short duration vehicle impact load to predict the coupler behavior, material properties and post impact performance as well. Splicesleeve along with high grade concrete grouted coupler are placed where plastic hinges are highly expected to form (Ebrahimpour et al. 2016), particularly at the pier and pierfoundation junction. Performance of the pier and the dynamic impact on it are studied for axially compressive stress, and the combined stresses as a result of residual flexure due to impact at column base for individual coupler. However, precise assessment of the performance of splicesleeve along with grouted coupler used in ABC needs deeper attention to be carried out (Ameli et al. 2016).
To analyze the impact characteristics of grouted coupler, it is placed in the pier foundation connection in order to evaluate the performance standard of the splice sleeve and grouted coupler mechanism as a composite material (Jacob et al. 2004). Coupler sections have been studied as a means of increasing the stiffness of the RC piers to improve their seismic performance and overcome the development of plastic hinges (Tazarv and Saiidi 2016; Thomas et al. 2018). In the present study, splicesleeves along with high grade concrete grouted coupler are embedded into the pierfoundation, placing the foundation top and the coupler crosssection in the same level. As such, failure mechanism of each coupler along with the material properties need rigorous prior investigation before recommending its widespread use in foundationpier connection for ABC under axially compressive stress, and the combined stresses resulting from residual flexure caused by impact load, transferring it at pier base (Zhou et al. 2017).
The importance of this holistic study is to examine static and dynamic characteristics of the single grouted coupler material used in RC ABC bridge pier to withstand short duration vehicle impact. In order to assess post impact behavior and enhancing the performance level, study of coupler on dynamic load requires additional scrutiny. In addition, to determine the stressstrain relationship of the coupler region and to directly capture stress and strain levels, some physical properties of the coupler system are studied comprising from uniaxial stress and strain system by utilizing the rebar stress and strain level.
2 Representative reinforced concrete (RC) pier
In this study, a representative circular reinforced concrete (RC) pier section is considered (as shown in Fig.Â 1). The concrete in the pier is specified to have a compressive strength of 3 ksi (20.68â€‰MPa) and the reinforcing steel a yield strength of 60 ksi (413.68â€‰MPa). The unrestrained length of the pier is taken as 8.6â€‰ft (2.62â€‰m) with circular crosssection throughout (Grouted splice sleeve connectors for ABC bridge joints in highseismic regions â€“ transportation blog n.d.) as shown in Fig. 1.The outer diameter and the inner diameter (h) of the concrete column are 21â€‰in. and 18â€‰in., respectively. The pier has primary reinforcement of 6 no. 8 ASTM 706 Grade 60 steel reinforcingbars and shear reinforcement of # 4 steel rebars at 2 \(\frac{1}{2}\) inches center to center (pitch) spirally conforming to the ACI minimum shear reinforcement criteria (ACI 2011). In addition, the representative pier also satisfies the shear reinforcement criteria for rebar diameter, and pitch of the spiral reinforcement (Furlong 2014).
3 Nominal moment strength (M_{N}) of reinforced concrete pier
Material and sectional details of the representative test column used in determining the nominal moment capacity are shown in TableÂ 1.
The detailed cross section with stress block diagram of the pier is further shown in Fig.Â 2. As per ACI specifications, a balanced strain condition is considered to exist at a crosssection where the tension steel reaches the strain corresponding to the yield strength of the steel (f_{y}) as the concrete crushing strain reaches its maximum value of 0.003. Thus, strain Îµ_{y} here, is determined as the yield strain of steel i.e. 0.00206 (f_{y} / E_{s}) (ACI 2011).
To determine the nominal moment capacity of the reinforced concrete section, a procedure as laid out by Hsiao is followed (Hsiao 2012). First, the distance of the extreme compression fiber to the neutral axis of the section is determined as shown in Eq. 1 (MacGregor et al. 2012).
Where: d_{t} is the distance from the extreme compression fiber to the extreme tension steel and calculated as; d_{t}â€‰=â€‰h â€“ cover â€“ diameter of main bar â€“ diameter of stirrup bar, and Îµ_{y} is the yield strain of the reinforcing steel.
The depth of the â€˜Whitneyâ€™ equivalent rectangular stress distribution in concrete for the balanced strain condition is then determined using Eq. 2.
Where: Î²_{1} is 0.85 for fâ€™_{c}â€‰â‰¤â€‰4000â€‰psi (27,579.03 kN/m^{2}) (ACI 2011).
The properties of circular compression block including its gross concrete area and centroid are then computed and as shown in the Eqs. 3 to 5 (Hsiao 2012).
Where: h is the diameter of the column, a_{b} is the depth of the â€˜Whitneyâ€™ block as determined in Eq. 2, A_{cb} is the area of the circular compression block and \(\overline{X}\) is the centroid location of the compression concrete block.
The computed area of the compression block is next used to determine the compression force in concrete for circular compression block, C_{c}, as shown in Eq. 6 (MacGregor et al. 2012).
3.1 Computation of the strains, and forces in tension and compression steel
Referring to Fig. 2 and from the geometry of similar triangles, strains in different fibers (Îµ_{s1}, Îµ_{s2}, and Îµ_{s3}) are computed, and using the computed values the forces in the steel (T_{1}, T_{2}, C_{s1} and C_{s2}) are determined as, T_{1}â€‰=â€‰A_{s} f_{y}, T_{2}â€‰=â€‰A_{s} *E_{s}*Îµ_{s3}, C_{s1}â€‰=â€‰A_{s} (f_{y} â€“ 0.85*fâ€™_{c}), and C_{s2}â€‰=â€‰A_{s} (Îµ_{s2}* E_{s} â€“ 0.85fâ€™_{c}).
The computed parameters are then used in calculating the nominal moment capacity (M_{n,g}) for the gross concrete area of the column (Fig. 1) as shown in Eq. 7 (Hsiao 2012).
From Eq. 7 and the specifications of the representative RC bridge pier, nominal strength of RC concrete for gross crosssectional area of column was computed as 3143.3236 kipin (385.65 kNm).
A similar process is used to estimate the nominal moment capacity of the core area of the column. Core concrete is confined by the laterally placed spiral tie to prevent lateral expansion due to Poissonâ€™s ratio, and incurs the axially compressed overloading due to impact. Confined concrete can enhance the performance in terms of capacity and deformability due to stress compatibility. The confinement of the core concrete contributes in improving ductility and after peakstress deformability in addition to increasing the axial load carrying capacity (Karim et al. 2014). The nominal core area capacity, computed using the reduced crosssectional area is determined to be 2007.206 kipin (226.78 kNm).
4 Determination of dynamic flexure
This study investigates sections of the vehicular impact scenario on reinforced concrete sections. A semitrailer is considered as the vehicle for the impact event as this represents a worsecase scenario. Vehicle weight and impact velocity of the semitrailer are considered as 80,000 lbs. (355.86 kN) and 100â€‰ft./sec (30.48â€‰m/sec) respectively to simulate a fully loaded trailer condition moving at its allowed speeds on the highway (Speeding and speed limits index and overview n.d.).
As a result of the parameters selected in this investigation, the vehicular impact scenario being studied is of the highvelocity, low duration variety. This will result in a high strain rate of loading for both the concrete and steel components of the RC pier. Concrete and steel both manifest a peculiar phenomenon when placed under such high loading rates (Auyeung et al. 2019). This phenomenon is an observable increase in strength capacity of these materials, as a function of the strain rate effect on reinforced concrete. As a result, a dynamic impact factor was proposed by (Malvar 1998), to reflect this increase on the strength capacity parameters of the reinforced concrete member.
4.1 Determination of dynamic increase factor (DIF)
Determination of the dynamic increase factor (DIF) reflecting the material behavior of concrete and steel during a vehicular impact scenario, involves an estimation of the dynamic flow stress from the impact. This dynamic flow stress can be estimated for either material. The dynamic flow stress (Ïƒ_{dyn}) in steel at impact is selected for use as it is projected to be the critical component of the member with respect to the coupler section. This dynamic flow stress is determined using Eq. 8 (Feyerabend 1988).
Where: Ïƒ_{y} is a static flow stress and is considered as 60 ksi (420â€‰MPa) for ASTM 706 Grade 60 steel rebar, C and p are the material constants (Cowper and Symonds 1957; Zhou and Li 2018) with values of 40 and 5 [8] respectively. Quasistatic strain rate of steel rebar (Î) is considered as 0.16â€‰sec^{âˆ’â€‰1} for impacting velocity 100â€‰ft./sec (30.48â€‰m/sec) (Cowper and Symonds 1957).
The dynamic parameter â€˜Î¾â€™ can be computed from Eq. 9 (Malvar and Crawford 1998; Mander et al. 1988) and Eq. 8.
Where: Î¾ is a constant which depends on the dynamic yield stress of steel at the strain hardening zone, Î is the strain rate of steel and Ïƒ_{dyn} is the dynamic flow stress at uniaxial plastic strain rate of steel.
The DIF can be computed from Eq. 10 (Malvar and Crawford 1998; Mander et al. 1988), using Eq. 9.
Replacing the value of Î¾ from Eq. 9 to Eq. 10, yields a DIF of 1.053.
4.2 Computation of static force at impact
The time dependent frontal shock from vehicular impact can be computed using an averaged integration of the instantaneous impact force over the range of 50â€‰ms near the peak impact force as shown in Eq. 11 (Zhou and Li 2018).
Where: I_{dyn} represents the frontal shock due to impact, I_{r,} is the peak reflected pressure (overpressure), t_{d}^{+} is the time instant of the peak impact force, and t represents the impact duration.
Developed to estimate the total static force of a vehicle impact from the instantaneous peak force occurring during impact, the relationship shown in Eq. 11 captures the expected loading history of the impact over time on the RC pier (Zhou et al. 2017) utilizing the expected sinusoidal loading pattern of the impact event to predict the dynamic load from the peak force and the loading time history.
The overpressure represented by I_{r,} is a function of the kinetic energy (E) from the impacting vehicle and can be determined using Eq. 12, Eq. 12 was developed as a relationship between bending stress developed in the pier from the peak dynamic force of impact and kinetic energy using data from various simulated and experimental studies (Cao et al. 2019; Gomez and Alipour 2014; Mohammed and Parvin 2013; Zhou and Li 2018). The bending stress was used in lieu of the impact force so as to capture the possible effects of geometric variations of the pier in the resulting overpressure from regression analysis at impact as shown in Eq. 12 (Roy et al. 2021). Boundary conditions of the pier is considered as bottom end fixed and top end restrained from displacement and rotation (Zhang et al. 2018) is as shown in Fig.Â 4c.
Where: E is the kinetic energy, absorbed by the impacted pier, I is the moment of inertia of the pier section, L is the height of the pier and c is the perpendicular distance from the neutral axis of the cross section to the farthest point on the cross section of the pier, as shown in Fig.Â 3.
Assuming the vehicle comes to rest without rebounding from the pier, the kinetic energy (E) equation is determined as same as the kinetic energy (E) of the vehicle using Eq. 13 (Tsang and Lam 2008).
Where: E is impact energy of the vehicle, M_{veh} represents the weight of the impacting vehicle, and V is the frontal impact velocity of the vehicle causing instability of the column.
4.3 Computation of external flexural moments
Moments caused by the lateral impact force from vehicle collisions are induced at the base of the column as well as at different levels between the point of impact and the base. These moments if exceeding the moment capacity of the column could result in structural failure.
Assuming a pinned connection for the RC column, the static moment (M_{s}) induced by the vehicular collision can be determined as shown in Eq. 14.
Where: H is the height (in feet) to compute static moment and I_{dyn} represents the frontal shock due to impact.
4.3.1 Dynamic moment
Dynamic moments (M_{dyn}) to investigate flexural effects at different fibers have been determined by static moment (M_{s}) times the dynamic increase factor (DIF). This is further expressed in Eq. 15 as to investigate the increased flexural effect (Feyerabend 1988).
Where: DIF is the dynamic impact factor and M_{s} is the static moment.
4.4 Unbalanced moments from static and dynamic impact
Unbalanced moments in the column are caused by the exceedance of the moment capacity of the column and by the moments induced by the impact force. These unbalanced moments can be resisted by the splicesleeve and grouted coupler system formed by the six couplers (six reinforcing steel bars are embedded into grout), one at the base of each longitudinal reinforcement as shown in the Fig. 4a. Determined as external static moments at column base for gross and core crosssectional area minus sectional moments of resistance at gross and core crosssectional areas, unbalanced static and dynamic moments for both gross and core crosssectional areas are computed as shown in Eqs. 19 through 22. FigureÂ 4a and b show the coupler arrangement in a column base embedded into the footing. Boundary conditions (fixfix ended at bottom and top end is restrained from displacement and rotation) are shown in Fig. 4c. Computations of unbalanced moments are computed while semitrailer impacts occur due collision at different heights in column has been further shown in Fig.Â 8.
4.4.1 Unbalanced static moment for gross crosssectional area
Static unbalanced moment (M_{un,sg}) for gross crosssectional area has been computed in Eq. 16, as,
Where: M_{s,g} is the external static moment for gross crosssectional area.
4.4.2 Unbalanced static moment for core concrete area
Static unbalanced moment (M_{un,sc}) for core crosssectional area has been computed in Eq. 17, as,
Where: M_{s,c} is the external static moment at core crosssectional area of concrete column.
4.4.3 Unbalanced dynamic moment for gross concrete area
Dynamic unbalanced moment (M_{un,dg}) for core crosssectional area has been computed in Eq. 18, as,
Where: M_{dyn,g} is the dynamic moment caused due to dynamic impact at gross crosssectional area of concrete pier.
4.4.4 Unbalanced dynamic moment for core concrete area
Dynamic unbalanced moment (M_{un,dc}) for core crosssectional area has been computed in Eq. 19, and as shown in Fig.Â 5.
Where: M_{dyn,c} is the dynamic moment caused due to impact at core crosssectional area of concrete.
5 Coupler detail to withstand unbalanced moments
Splicesleeve used for grouted coupler in the column rebar properly embedded and placed (Fig. 4) appreciably predicts better performance in dynamic response. Grouted couplers are embedded in the foundation to act as a rigid body and in conjunction with the foundation, attract the maximum impact force. Foundations are rigid bodies and are less slender than piers. So a coupler embedded in the foundation is expected to perform better than if embedded in the pier (Pantelides et al. 2014). For this particular study, the splice sleeve detail recommended as per â€˜Splice Sleeve North Americaâ€™ and â€˜NMB General Brochureâ€™ (ICCES Evaluation Report ESR3433 2014) are considered and shown in Fig.Â 6.
For this particular study, sleeve number 8â€‰UX (ICCES Report 2016) has been selected for the # 8 ASTM 706 (ASTM 2015) bars used in column for main reinforcement. Splicesleeve (ICCES Report 2016) details used in this study are tabulated and shown in TableÂ 2.
Various connection types have been studied for precast concrete bridge columns in seismic areas in two major categories of emulative and rocking connections (GirÃ£o Coelho et al. 2012). The emulative connection for precast components is defined as a connection that incorporates special details resulting in a performance that emulates that of a monolithic castinplace component as shown in the Fig.Â 7. The steel rebar becomes discontinuous at the middle of splice sleeve.
5.1 Computation of bending stresses at column base
In order to estimate the bending stresses occurring at the column base due to the unbalanced moment from the impact, some geometric properties of the column such as the distance of the extreme fiber from the neutral axis (NA) as well as the moment of inertia need to be evaluated. Determination of bending (flexural) stresses (Ïƒ_{b}) experienced by the pier is as shown in Eq. 20 (Ameli and Pantelides 2017).
Where: M_{un} represents the unbalanced moments for static, dynamic and for both gross and core crosssectional areas (Fig. 4). â€˜câ€™ represents the distance of the extreme fiber from in the neutral axis (as shown in Fig. 3), and I_{avg,} is the average moment of inertia of the column cross section.
5.2 Determination of direct stresses
Direct stress (Ïƒ_{o}) and design axial compressive load the pier undertaken have been computed from the Eq. 21 (Ameli and Pantelides 2017) and 22 (MacGregor et al. 2012), respectively.
Where: fâ€™_{c} is the 28day concrete compressive strength, A_{g} is the gross crosssectional area, A_{st} is the area of 6 # 8 (1â€‰in. or 25.4â€‰mm diameter) reinforcing bars, and Ïƒ_{y} is the yield strength of the reinforcing steel rebar.
Details of the necessary input data for the stress computations are tabulated in TableÂ 3, and the resulting stresses in TableÂ 4.
5.3 Determination of combined stresses governed by grouted coupler
The stress experienced by each coupler (Ïƒ) is expressed as the sum of direct stress (Ïƒ_{o}) and bending stress (Ïƒ_{b}) (GirÃ£o Coelho et al. 2012). Known as combined stress, this summation of stresses is computed as shown in Eq. 23.
Where: Ïƒ_{o} and Ïƒ_{b} represent direct and bending stress respectively.
In Eq. 25, (+) and (âˆ’) signs indicate tensile and compressive stresses, respectively.
To analyze the response of the reinforced concrete pier to external impact events, two scenarios are studied. These are the effect of impact at different heights on the resulting moments and stresses at the base of the column as well as the resulting moments and stresses at different points of the pier from impact at a particular point.
5.3.1 Case â€“ I Impact at different heights
Computation of combined stresses at the base in the column for single coupler caused by the impacts in column are shown in Fig. 7. Different impact heights are considered as the frontal height of the semitrailer is approximately 3â€‰ft. (1â€‰m). Maximum height of impact is taken as 4.029â€‰ft. (1.5â€‰m) to account for possible elevation of the semitrailer bouncing off a bumper just prior to impact. For each single vehicular hit, combined stresses (both tensile and compressive) experienced by the column and by extension each coupler are shown in Fig. 8.
From the Fig. 9, it can be deduced that there is a linear trend in the stresses with a change in height of impact location. The stresses at the base increase as the height of the impact location increases. This can be attributed to the increase in the moment generated at the base with an increase in the distance of the impact from the base. Also, there is very little difference in the stresses occurring in the gross and core areas. This means that the concrete cover adds only a very limited capacity to the overall capacity of the column. However, there is a distinct change in the resulting stresses when the effect of strain rate is included using the DIF. The resulting stresses (labeled as dynamic stresses) are significantly larger than those from a static loading scenario, indicating that the use of static forces without adequately compensating for the dynamic nature of the impact loads will lead to overly liberal estimates of the stresses.
5.3.2 Case â€“ II Stresses at different heights from impact at a single location
Combined stresses at different fiber levels from impact at a constant height of 3â€‰ft is as shown in Fig. 9. Direct stress from induced from axially compression load and flexural stress from bending caused due to impact are summed up to compute the resultant combined stresses for both static and dynamic are as shown in Figs.Â 10 and 11.
From the results, it can be observed that although the maximum balanced moments, and by extension the maximum combined stresses, occur at the base of the column (i.e. at the maximum distance from the point of impact), moment overpressures also occur at other sections of the column. These moment overpressures result in combined stresses exceeding the capacity of the pier. This implies that although the use of couplers in the base of the pier may improve its stiffness and thus its resistance to unbalanced moments at the base, the column will still possibly fail at other locations away from the base where the coupler has no significant effect. As a result, in addition to the use of couplers, other measures need to be adopted to improve the moment capacity of the pier and prevent collapse.
6 Finite element model (FEM) of the grouted coupler
To corroborate the results obtained from the analytical analyses carried out, a composite finite element model (FEM) consisting of concrete cylinder (3 ksi or 20.68â€‰MPa), cast iron hollow cylindrical splice sleeve, concrete grout (6 ksi or 41.36â€‰MPa), and single reinforcing steel reinforcing bar are designed and analyzed using the commercial software ANSYS Mechanical. This model is designed to characterize a single coupler unit with concrete cover. Hollow cylindrical cast iron splicesleeve (36 ksi or 248â€‰MPa)) is used along steel rebar (60 ksi or 413â€‰MPa) embedded in the splice sleeve. FigureÂ 12 shows the designed finite element model (FEM) of the grouted coupler. For all three different materials and their attachments, nonseparable contacts have been considered to act as a monolithic behavior of the model under vertical axial compression and horizontal impact. During simulation, high frictions are developed at the contacts of all intermaterial surfaces. The couplerrebar model considered in this study shows large deformation as a result of transmitting horizontal load while the RC bridge pier experiences vehicle impact. Material properties utilized for developing the numerical model are shown in TableÂ 5.
A rectangular mesh is applied to the model resulting in 7080 elements. FigureÂ 13 shows the resulting meshed model.
Axial load from the superstructure is scaled down with respect to the crosssectional area of the model. The load from impact is applied in a similar version albeit as a moment occurring at the base of the column.
7 Results from FEM
Analyses of the model under both static and dynamic loading conditions are performed. Results from these analyses are presented in this section.
7.1 Stress concentration in coupler
FiguresÂ 14 and 15 show the resulting combined axial and bending stresses from the static and dynamic loading conditions respectively. The results show an agreement in the values of maximum stresses obtained from the calculations. The maximum stresses from the finite element analyses are slightly higher than the those obtained from corresponding stress calculations. This can be attributed to an increased precision expected of finite element analysis in comparison to less complicated computation methods.
Figures 14b and 15b show stress concentration from the applied loads within the grouted coupler. This indicates that the coupler system bears the brunt of the unbalanced moment resulting from the impact scenario. This leads to an extensive crack system developing in the concrete grout inside the splice sleeve. FiguresÂ 16 and 17 show these crack patterns.
From Figs.Â 16 and 17, it can be deduced that in resisting most of the unbalanced moment, the coupler system ensures the concrete column retains its serviceability. The crack pattern shows that all resulting cracks from the loading developed in the concrete grout, limiting damage to the column in just the coupler system, and leaving the surrounding concrete intact and thus serviceable.
7.2 Strain in coupler
FigureÂ 18 shows the strain developed in the coupler in the finite element model. This coupler model, corresponding to a length ratio of 1 (only a section of the column corresponding to the length of the coupler was modeled), with a rigid length factor of 0.65 (from the proposed value for grouted couplers in (Tazarv and Saiidi 2016)).
8 Uniaxial stressstrain concentration model for coupler
Spliced bars used as grouted couplers exhibit different behavior to reinforcing steel reinforcing bars due to the anchoring mechanism (bond behavior of coupler system with reinforcement) (Tazarv and Saiidi 2016). The anchoring mechanism for the spliced bar manifests a rigid body behavior depending on the stressstrain relationship and the stress concentration at the coupler region (L_{critical}). To determine the stressstrain relationship of the coupler region some physical properties of the coupler system, including the coupler rigid length factor (Î²) and its length play a significant role. Under tension, only a part of the coupler system will undergo elongation. A section in the middle will stay rigid due to larger diameter of the coupler relative to the reinforcing steel as well as its anchoring mechanism (Tazarv and Saiidi 2016). This rigid length can be determined using a coupler rigid length factor (Î²) which is a function of the coupler type. This factor has been experimentally determined for different coupler types in (Tazarv and Saiidi 2016). Strain in the coupler can be computed using the relationship in Eq. 24 (Tazarv and Saiidi 2016).
Where: Îµ_{sp} is the required strain at the coupler region, Îµ_{y} is the strain at the connecting steel rebar, L_{sp} is the coupler length, L_{critical} is the coupler region length, and b is the coupler rigid length factor.
FigureÂ 19 shows the geometry of the coupler system with some of its physical dimensions, which are used in Eq. 24.
The coupler region extends beyond each side of the coupler sleeve length (L_{sp}), by a factor defined by the diameter of reinforcing steel (d_{b}) due to its having a higher stiffness than the adjacent steel rebar as shown in Fig.Â 18 (Haber et al. 2014). The coupler critical length (L_{critical}) can be computed using its physical length (L_{sp}) and the diameter of the reinforcing steel as shown in Eq. 25 (Tazarv and Saiidi 2016).
Where: d_{b} is the diameter of reinforcing steel rebar, L_{sp} is the coupler sleeve length, L_{critical} is the coupler critical length, and Îº is the factor varies from 1.0 to 2.0 (Ameli and Pantelides 2017).
In order to determine the influence area of the highly stressed zone induced by impact, L_{critical} has been given substantial importance, and needs to be computed. The length L_{critical} is actually extended beyond the splicesleeve zone and some portions of the steel rebar extended from both the sides are considered as shown in Fig. 19. To understand the deformations at steel rebar and splicesleeve junction, Îº plays a very sensitive role for the computation of the highly stressed zone, the critical length (L_{critical}) of the coupler.
FigureÂ 20 shows the linear relationship used to identify L_{critical} from the defined coupler sleeve length (L_{sp}). In order to determine L_{critical} from the specified L_{sp} carried out from the conservative Îº values ranging from 1.0 to 2.0 (Ameli and Pantelides 2017), Ï‡, the ratio of coupler length (L_{sp}) over the coupler region length (L_{critical}), can be computed using Eq. 25 and Fig.Â 18. L_{sp} data can be taken from the splice sleeve specification (ICCES Evaluation Report ESR3433 2014). In this Ï‡ is very significant to compute Îµ_{sp} as mentioned in Eq. 27. FigureÂ 18 shows the relationship between L_{critical} and L_{sp} to in order to establish a holistic methodology for the different Îº factors ranging from 1.0 to 2.0. This method includes coupler length (L_{sp}) to determine maximum strain developed at the coupler and steel rebar region while experiencing vehicle impact. In order to ascertaining static strain (Îµ_{sp}) at coupler region, Î² and Ï‡ play an important role for the chronological computation of Îµ_{sp}, which are shown in Eqs. 26 and 27.
Modification and rearrangement of Eqs. 24 and 25, yields Eq. 26.
In Eq. 26, Ï‡, the ratio of coupler length over the coupler region length (L_{sp}/L_{critical}), Î² is the coupler rigid length factor (varies from 0.0 to 1.0) are introduced to predict coupler region induced strain (Îµ_{sp}). Furthermore, to predict the strain at coupler region, Îµ_{sp} can be determined from Eq. 27 after rearranging the Eq. 26.
From Eq. 27, it can be deduced that the strain in the coupler system is a function of its geometric parameters as well as the strain induced in the reinforcing steel. As a result, the performance of the coupler system can be designed to meet desired specifications by modifying some of the geometric parameters of the coupler.
From Eq. 26, coupler strain (Îµ_{sp}) is a function of the mechanical properties of the coupler, including its rigid length factor and its region length.
A maximum strain of 0.00076 is developed in the coupler from the simulated impact occurring on the concrete pier. This is identical to the expected value of 0.00073 computed using Eq. 27, indicating that the equation provides a good estimate of the relationship between a couplerâ€™s mechanical properties and the strain expected to occur in it from external loads.
Considering the modulus of steel reinforcement (E_{s}) as 29*10^{6}â€‰psi, and yield stress of steel rebar (Ïƒ_{y}) as 60 ksi, the strain in the steel rebar (Îµ_{y}) is computed using Hookeâ€™s law as 0.0021. The relationship between the mechanical properties of the coupler and the resulting strain is shown in Fig.Â 21. This figure plots the expected coupler strain for various coupler types (identified by the different rigid length factors Î²) at different length ratios (Ï‡).
Coupler strains (Îµ_{sp}) for different coupler length ratio (Ï‡) ranging from 0.625 to 0.80 and at the corresponding Îº values (ranging between 1.0 and 2.0) are computed for steel strain of 0.0021. These plots allow for an estimation of the strain that will be developed in a coupler system with respect to its mechanical properties. Each plot represents a specific coupler rigid length factor (Î²), which is a function of the coupler type. Respective coupler strains are computed for Î² ranging between 0 to 1, and as shown in Fig. 21. The plots show a linear relationship between the coupler strain and its length ratio. This relationship is more evident at higher Î² values and trends towards a uniform strain value at the Î² of zero. This indicates that the material and geometrical properties of the coupler play an important role in its strain capacity, helping in designing coupler systems to meet specific service demands.
8.1 Determination of strainrate at coupler region
During dynamic impact, the strain rate in the critical region (L_{critical}) changes linearly with changes in the strain rate of the steel rebar connected to the coupler. Strain rates of the steel rebars due to vehicle impact on a circular RC bridge pier are estimated to range between 10^{âˆ’â€‰1} and 10^{âˆ’â€‰4} for quasistatic to dynamic state at a yield stress of 60 ksi (Malvar and Crawford 1998). Dynamic strain rates in the coupler region can be assessed by partial derivative of Eq. 27 with respect to time (t) in seconds. This yields Eq. 28 which can be used to predict strain rates in the coupler region. Steel rebar strain rate (Î´Îµ_{y}/Î´t) and strain rate at coupler (Î´Îµ_{sp}/Î´t) are expressed as Î_{y} at yield as Î_{sp} in Eq. 28.
Where: The variables Î_{y,}Î_{sp,}Î², and Ï‡ are already explained.
Simulations of the expected coupler strain rates (Î_{sp}) at different coupler length ratios (Ï‡) and for the different steel rebar strain rates are investigated. This could feasibly occur during the impact events are undertaken to gain an insight into the behavior of the coupler at changing strain rates in steel rebar. Shown in the Figs.Â 22, 23 and 24 are the respective different steel rebar strain rates of 10^{âˆ’â€‰1}â€‰sec, 10^{âˆ’â€‰2}â€‰sec, 10^{âˆ’â€‰3}â€‰sec, and 10^{âˆ’â€‰4}â€‰sec, for increasing Ï‡ values at the specified coupler strain rates depending on various Îº factors ranging from 1 to 2.
FiguresÂ 22, 23, 24 and 25 show the different strain rates expected in the coupler region from various impact conditions at different coupler length ratios (Ï‡) and at different possible steel rebar strain rates (Î_{y}). Coupler region strain rates (Î_{sp}) show the linear trends with respect to the Ï‡ values, indicating an increase in the expected strain rate in the coupler region with a reduction in its critical length. As such, for the known strain rates in steel, which is a function of loading, the corresponding coupler strain rates can be easily determined.
9 Discussions and conclusions
In reinforced concrete (RC) structures, piers are usually the most vulnerable members to collisions due to its exposed face and slender behavior. In particular, the desired characteristics and the associated impact performance levels of the critical components warrant additional investigation. The following observations and conclusions are drawn from this study:

In this research, a means of addressing a critical component of the pier behavior during a vehicular impact scenario is investigated. This involves the use of a grouted coupler connection system to improve stiffness, and thus increase the moment capacity of the pier and by extension reduce the risk of collapse of the pier from unbalanced moments expected to occur during impact. Combined stresses in the base and at different fiber sections as a result of vehicular impact in the pier are analyzed, and the results are shown for both bending compression and bending tension. A finite element model (FEM) is also used to corroborate the computed stresses expected to occur from vehicle impact at the base of the pier, as well as to investigate the benefit of including a grouted coupler sleeve at this vulnerable spot to limit the damage taking place during impact. The investigations reveal that the grouted coupler performs quite well in containing the unbalanced moments from the impact, without cracking the concrete of the pier footing, while observing cracks in the grout for both static and dynamic impact loads.

Further investigations into the strain rate developed in the coupler during impact provide an insightful notion in establishing and understanding the relationship between strain and mechanical properties of the coupler at impact load. This relationship ensures that couplers going forward can be designed to adequately compensate for loading scenarios resulting in moments, exceeding the design capacity of the reinforced concrete (RC) bridge piers, and thus save them from excessive damage. Beyond the investigations carried out in this study, it is also imperative to investigate the behavior of grouted couplers in other positions besides being embedded in the foundation. Also, other possible materials for the coupler sleeve besides cast iron as well as some other grout mixes (with higher compressive strengths) need to be scrutinized to fully comprehend the influence of the coupler system to a pierâ€™s resistance in counteracting impact loads.

This study is an attempt to determine the dynamic strain comprising the model during and post impact scenario. Maximum strain can be found at Î²â€‰=â€‰0, which corresponds to having no coupler and the entire strain developed is totally induced in the steel rebar. The strain rate is particularly useful for very short impact durations where instantaneous strain is impossible to determine.
The methods utilized in this study can also help improve design, providing designers with information on how the grouted couplers can be expected to behave under lateral impact loads. However, further experimental studies involving various pier geometries, different material properties, and changing the couplerâ€™s position are warranted to investigate and capture the performance under various impact scenarios.
Availability of data and materials
Some or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.
Abbreviations
 M_{n} :

Nominal moment strength
 E_{s} :

Modulus of elasticity of steel
 E_{c} :

Modulus of elasticity of concrete
 fâ€™_{c} :

Compressive strength of concrete
 f_{y} :

Yield strength of steel
 Îµ_{y} :

Strain at steel corresponding to f_{y}
 c:

Height of compression block
 n:

Modular ratio
 Î²_{1} :

Ratio of depth of rectangular stress block to the depth to the neutral axis
 h:

External diameter of the concrete column
 d_{t} :

Distance from extreme compression fiber to extreme tension steel
 x_{b} :

Distance from the extreme compression fiber to the neutral axis for a balanced strain condition
 a_{b} :

Depth of Whitney equivalent rectangular stress distribution in concrete for a balanced strain condition
 Î±:

Angle between two rebarâ€™s at center of the circular area
 A_{cb} :

Area of circular compression block
 \(\overline{\mathrm{X}}\) :

Centroid location of circular compression block
 C_{c} :

Compressive force in concrete for circular compression block
 C_{s} :

Compressive force in longitudinal reinforcement
 T:

Tensile force of steel
 P_{n} :

Nominal axial compressive strength
 P_{a} :

Axial compressive strength
 M_{n,g} :

Nominal moment capacity for the gross concrete area of the column
 h_{c} :

Diameter of the core concrete
 A_{gc} :

Gross area of the core concrete
 A_{nc} :

Net area of the core concrete
 A_{c} :

Area of the core concrete
 A_{g} :

Gross crosssectional area
 C_{c,c} :

Compressive for of the core concrete
 P_{n,c} :

Nominal axial compressive strength of the core concrete
 M_{n,c} :

Nominal flexural capacity for the core concrete section
 Ïƒ_{dyn} :

Dynamic flow stress
 Î:

Quasistatic strain rate of steel rebar
 Î_{y} :

Quasistatic strain rate at steel rebar
 Î_{sp} :

Strain rate at coupler
 Ïƒ_{y} :

Static flow stress
 Ïƒ_{dyn} :

Dynamic flow stress in steel during impact
 Î¾:

Dynamic parameter
 DIF:

Dynamic increase factor
 I_{dyn} :

Frontal shock due to impact
 I_{r,} :

Peak reflected pressure (overpressure)
 t_{d} ^{+} :

Time instant of the peak impact force
 t:

Impact duration
 V:

Maximum permissible velocity
 M_{s} :

Static moment
 t :

Time of collision
 H:

Height of collision
 M_{dyn} :

Dynamic moment
 M_{un,sg} :

Unbalanced moment for static gross crosssectional area
 M_{s,g} :

Static moment at gross concrete area
 M_{n,g} :

Nominal moment capacity for gross concrete area
 M_{un,sc} :

Static unbalanced moment for core crosssectional area
 M_{un,dg} :

Dynamic unbalanced moment for core crosssectional area
 M_{dyn,g} :

Dynamic moment at gross concrete area
 M_{un,dc} :

Dynamic unbalanced moment for core crosssectional area
 Ïƒ_{b} :

Bending stress
 Ïƒ_{o} :

Direct stress
 I_{avg} :

Average moment of inertia of each splice sleeve
 c :

Distance of the extreme fiber from the centroid
 P :

Axial compressive force
 Ïƒ:

Combined stress
 W:

Wider end of the splice sleeve
 N:

Narrower end of the splice sleeve
 Ï‡:

Coupler length ratio (L_{sp} / L_{cr})
 Î²:

Coupler rigid length factor
 L _{ sp } :

Coupler length
 L _{ critical } :

Coupler region length
 Îº:

A factor ranges from 1.0 to 2.0
References
ACI (2011) ACI 31811: building code requirements for structural concrete. American Concrete Institute,Â United States of America
Ameli MJ, Brown DN, Parks JE, Pantelides CP (2016) Seismic columntofooting connections using grouted splice sleeves. ACI Struct J 113(5):1021â€“1030 American Concrete Institute
Ameli MJ, Pantelides CP (2017) Seismic analysis of precast concrete bridge columns connected with grouted splice sleeve connectors. J Struct Eng 143(2):4016176 American Society of Civil Engineers
ASTM (2015) A706/A706M âˆ’ 15 standard specification for lowalloy steel deformed and plain bars for concrete. ASTM International,Â United States of America, pp 1â€“6
Auyeung S, Alipour A, Saini D (2019) Performancebased design of bridge piers under vehicle collision. Eng Struct 191:752â€“765 Elsevier
Cao R, Agrawal AK, ElTawil S, Xu X, Wong W (2019) Heavy truck collision with bridge piers: computational simulation study. J Bridge Eng 24(6):4019052 American Society of Civil Engineers
Cowper G, Symonds P (1957) Strain hardening and strainrate effects in the impact loading of cantilever beam. Brown University Division of Applied Mathematics, United States of America, pp 1â€“46
Ebrahimpour A, Earles BE, Maskey S, Tangarife M, Sorensen AD (2016) Seismic performance of columns with grouted couplers in Idaho accelerated bridge construction applications. Transportation Dept, Idaho
ElTawil S, Severino E, Fonseca P (2005) Vehicle collision with bridge piers. J Bridg Eng 10(3):345â€“353
Feyerabend M (1988) Hard transverse impacts on steel beams and reinforced concrete beams. University of Karlsruhe (TH), Germany
Furlong RW (2014) Design for shear, p 37
GirÃ£o Coelho AM, SimÃ£o PD, Bijlaard FSK (2012) Guidance for the design of spliced columns. J Struct Eng 138(9):1079â€“1088 American Society of Civil Engineers
Gomez NL, Alipour A (2014) Study of circular reinforced concrete bridge piers subjected to vehicular collisions. Struct Congress 2014:577â€“587
Grouted splice sleeve connectors for ABC bridge joints in highseismic regions â€“ transportation blog (n.d.). https://blog.udot.utah.gov/2014/09/groutedsplicesleeveconnectorsforabcbridgejointsinhighseismicregions/. (Jul. 13, 2020)
Haber ZB, Saiidi MS, Sanders DH (2014) Seismic performance of precast columns with mechanically spliced columnfooting connections. ACI Struct J 111(3):639â€“650 American Concrete Institute
Hsiao JK (2012) Bendingaxis effects on loadmoment (PM) interaction diagrams for circular concrete columns using a limited number of longitudinal reinforcing bars. Electron J Struct Eng, EJSE International Ltd., University of Melbourne Grattan St. Parkville VIC 12:1
ICCES Evaluation Report ESR3433 (2014). https://icces.org/reportlisting/esr3433/
ICCES Report (2016).Â https://icces.org/reportlisting/esr3433/
Jacob GC, Fellers JF, Starbuck JM, Simunovic S (2004) Crashworthiness of automotive composite material systems
Karim H, Sheikh MN, Hadi MNS (2014) Confinement of circular concrete columns: a review. In: Proceeding of The 1st International Engineering Conference on Developments in Civil & Computer Engineering Applications (IEC2014), pp 28â€“36
MacGregor JG, Wight JK, Teng S, Irawan P (2012) Reinforced concrete: mechanics and design. Prentice Hall, Upper Saddle River
Malvar LJ (1998) Review of static and dynamic properties of steel reinforcing bars. ACI Mater J 95(5):609â€“616
Malvar LJ, Crawford JE (1998) Dynamic increase factors for steel reinforcing bars, 28th DDESB seminar edn, Orlando.Â https://www.researchgate.net/publication/235099732_Dynamic_Increase_Factors_for_Steel_Reinforcing_Bars
Mander JB, Priestley MJ, Park R (1988) Theoretical stressstrain model for confined concrete. J Struct Eng (United States) 114(8):1804â€“1826
Mohammed TA, Parvin A (2013) Evaluating damage scale model of concrete materials using test data. Adv Concrete Constr 1(4):289â€“304 TechnoPress
Pantelides CP, Ameli MJ, Parks JE, Brown DN (2014) Seismic evaluation of grouted splice sleeve connections for precast RC bridge piers in ABC. Utah Department of Transportation,Â United States of America
Roy S, Sorensen A (2021a) Energy based model of vehicle impacted reinforced bridge piers accounting for concrete contribution to resilience. 18th international probabilistic workshop: IPW 2020. Springer Nature, United States of America, p 301
Roy S, Sorensen A (2021b) A reliability based crack propagation model for reinforced concrete bridge piers subject to vehicle impact. 18th international probabilistic workshop: IPW 2020. Springer Nature, United States of America, p 95
Roy S, Unobe I, Sorensen AD (2021) Vehicleimpact damage of reinforced concrete bridge piers: a stateofthe art review. J Perform Constr Facil 35(5):03121001 American Society of Civil Engineers
Roy S, Unobe ID, Sorensen AD (2022) Reliability assessment and sensitivity analysis of vehicle impacted reinforced concrete circular bridge piers. Structures 37:600â€“612 Elsevier
Sharma H, Gardoni P, Hurlebaus S (2015) Performancebased probabilistic capacity models and fragility estimates for RC columns subject to vehicle collision. ComputAided Civ Infrastruct Eng 30(7):555â€“569
Speeding & speed limits index & overview (n.d.). https://www.dot.nd.gov/divisions/programming/docs/SpeedLimitGuidelines.pdf. Accessed September 2015
Tazarv M, Saiidi MS (2016) Seismic design of bridge columns incorporating mechanical bar splices in plastic hinge regions. Eng Struct 124:507â€“520 Elsevier
Thilakarathna HMI, Thambiratnam DP, Dhanasekar M, Perera N (2010) Numerical simulation of axially loaded concrete columns under transverse impact and vulnerability assessment. Int J Impact Eng 37(11):1100â€“1112 Pergamon
Thomas RJ, Steel K, Sorensen AD (2018) Reliability analysis of circular reinforced concrete columns subject to sequential vehicular impact and blast loading. Eng Struct 168:838â€“851 Elsevier
Tsang H, Lam NTK (2008) Collapse of reinforced concrete column by vehicle impact. ComputAided Civil Infrastruct Eng 23(6):427â€“436 Wiley Online Library
Zhang G, Chen Z, Lu J, Xu S, Zhou X (2018) Experimental study on the impact properties of concrete bridge pier reinforced with stainless steel rebar. J Test Eval 46(4):1650â€“1658 ASTM International
Zhou D, Li R (2018) Damage assessment of bridge piers subjected to vehicle collision. Adv Struct Eng 21(15):2270â€“2281 SAGE Publications Sage UK: London, England
Zhou D, Li R, Wang J, Guo C (2017) Study on impact behavior and impact force of bridge pier subjected to vehicle collision. Shock Vib 2017:1â€“12
Acknowledgements
This publication was supported by a subcontract from Rutgers University, Center for Advanced Infrastructure and Transportation (CAIT), under DTFH6208C00005 from the U.S. Department of TransportationFederal Highway Administration (USDOTFHWA) and NMB SpliceSleeve. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of Rutgers University and NMB SpliceSleeve or those of the U.S. Department of TransportationFederal Highway Administration.
Funding
1) Center for Advanced Infrastructure and Transportation (CAIT), under DTFH6208C00005 from the U.S. Department of TransportationFederal Highway Administration (USDOTFHWA).
2) NMB SpliceSleeve â€“ North America, USA.
Author information
Authors and Affiliations
Contributions
Suman Roy^{*1}: Conceptualized and developed the theoretical formalism, numerical models (FEM) and simulations, validation, performed the analytic calculations, and writing draft. Ikwulono D. Unobe^{1}: Apprehended the analytic calculations, performed the numerical simulations, and edited the draft. Andrew D. Sorensen^{1}: Conceptualized and overall guidance, editing manuscript and supervised the project. The authors have read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
We, the authors of the manuscript declare that we have no competing interests.
Additional information
Publisherâ€™s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Table 6
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Roy, S., Unobe, I.D. & Sorensen, A.D. Investigation of the performance of grouted couplers in vehicle impacted reinforced concrete ABC bridge piers. ABEN 3, 18 (2022). https://doi.org/10.1186/s4325102200065y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s4325102200065y
Keywords
 Grouted coupler
 Splice sleeve
 Bridge pier
 Unbalanced moments
 Impact
 Strain